Introducing: Ranked Ladder 2.0
[h3]A new approach to the calculation of player rating in Gremlins, Inc.[/h3]
The current formula that calculates player rating in ranked sessions is in the game for several years by now. We recently went through all the player feedback regarding ranked sessions that we received over the years, and decided that it’s high time to implement a rather radical change. Today we’d like to present our ideas to you and to discuss it in the time that is still left before we introduce the change with the Autumn update.
How does the rating system work now?
An additional nuance is that difference in rating between 2 players also makes a difference:
If you “win” against a player with a higher rating than yours (you scored higher than them in the session), you get more rating points from them. If they “win” against you, they get fewer rating points from you. However, this additional rule isn’t significant enough to, for example, enable the player in the 3rd place to finish with a positive rating, no matter who are their opponents.
So this is the formula that the game relied on for almost 5 years by now.
And here are its main drawbacks:
The relative position at the end of the session is more important than the victory
At its core, the system rewards all places in the session except for the last one.
A player can finish in the 2nd place and technically “lose” a session (because someone else has won it) but still increase their overall rating – which is considered as “victory” by many players.
This duality of the outcome does not encourage the play style “victory at any cost”. Instead, it encourages to play for the highest place that the player deems possible for themselves, because victory is good – but in no way necessary for the progression through the ranked ladder. A higher position guarantees a better result on its own, even if it is 3rd place versus 4th place.
It is possible to mitigate the loss of rating points – but this happens at the expense of other losing players instead of the leader
In order to save your own rating, sometimes you have to ignore the leader and instead join them in their attacks on someone else. There are cases when this is the right thing to do, and finding that moment of “betrayal” in a session requires skill and impeccable timing! But in many cases, poorly implemented strategy of the protection of one’s own position leads to nothing but a sub-par experience for other players.
This approach goes against the spirit of most competitive board games
In most competitive board games, anything but the first place is considered to be a loss. The current ranked formula of Gremlins, Inc., where “playing for 2nd place” is a viable strategy, is seen by some board game players as “unfair” and “unsportsmanlike”.
[h3]Changes that we plan to introduce[/h3]
We believe that the new formula should emphasize victory and playstyle directed toward the 1st place. After all, Gremlins, Inc. is a game about prestige. For gremlins, there is no higher goal than accumulating a lot of prestige – represented by player score. The player with the highest score wins the session, everyone else loses – and the change in rating should reflect this.
Main principles:
Rating loss by each loser = ( ːscoreː of the player - ːscoreː of the winner) / ːscoreː of the winner * 4 / number of players in the session. Rating gain by the winner = the sum of all rating losses.
[h3]What do we want to achieve with this change?[/h3]
[h3]What are the possible issues with the new system?[/h3]
As you can see, there are pros that can be very beneficial for the gameplay in ranked sessions, but there are also cons that have to be addressed in order to keep the game fair and interesting.
We definitely want to try the new formula for at least a few months, but we also are very interested to know your opinion about such a change: please share your comments here on Steam, or on the game’s Discord server, so that together we can create a better player experience.
Thanks for staying with us, and we look forward to the community discussion!
______
ːmaliceː Official Discord
ːmaliceː Twitter (game updates)
ːsummer_magicː Official Discord
ːsummer_magicː Twitter (game updates)
The current formula that calculates player rating in ranked sessions is in the game for several years by now. We recently went through all the player feedback regarding ranked sessions that we received over the years, and decided that it’s high time to implement a rather radical change. Today we’d like to present our ideas to you and to discuss it in the time that is still left before we introduce the change with the Autumn update.
How does the rating system work now?
- When a ranked session ends, each participant finishes it at a particular place (1st, 2nd, etc.).
- Each player “takes” rating points from every opponent who has finished at a lower place than them.
- Player at the 1st place takes points from 3 other players (if we look at 4-player ranked sessions on the European server). This player gets the most rating points.
- Player at the 2nd place takes rating points from those who occupy 3rd and 4th places, while giving some of their own points to the player who took the 1st place. The exchange is almost always good enough for the player at the 2nd place to increase their rating points.
- Players in 3rd and 4th places lose rating points, because they give away more than they take from others.
An additional nuance is that difference in rating between 2 players also makes a difference:
If you “win” against a player with a higher rating than yours (you scored higher than them in the session), you get more rating points from them. If they “win” against you, they get fewer rating points from you. However, this additional rule isn’t significant enough to, for example, enable the player in the 3rd place to finish with a positive rating, no matter who are their opponents.
So this is the formula that the game relied on for almost 5 years by now.
And here are its main drawbacks:
The relative position at the end of the session is more important than the victory
At its core, the system rewards all places in the session except for the last one.
A player can finish in the 2nd place and technically “lose” a session (because someone else has won it) but still increase their overall rating – which is considered as “victory” by many players.
This duality of the outcome does not encourage the play style “victory at any cost”. Instead, it encourages to play for the highest place that the player deems possible for themselves, because victory is good – but in no way necessary for the progression through the ranked ladder. A higher position guarantees a better result on its own, even if it is 3rd place versus 4th place.
It is possible to mitigate the loss of rating points – but this happens at the expense of other losing players instead of the leader
In order to save your own rating, sometimes you have to ignore the leader and instead join them in their attacks on someone else. There are cases when this is the right thing to do, and finding that moment of “betrayal” in a session requires skill and impeccable timing! But in many cases, poorly implemented strategy of the protection of one’s own position leads to nothing but a sub-par experience for other players.
This approach goes against the spirit of most competitive board games
In most competitive board games, anything but the first place is considered to be a loss. The current ranked formula of Gremlins, Inc., where “playing for 2nd place” is a viable strategy, is seen by some board game players as “unfair” and “unsportsmanlike”.
[h3]Changes that we plan to introduce[/h3]
We believe that the new formula should emphasize victory and playstyle directed toward the 1st place. After all, Gremlins, Inc. is a game about prestige. For gremlins, there is no higher goal than accumulating a lot of prestige – represented by player score. The player with the highest score wins the session, everyone else loses – and the change in rating should reflect this.
Main principles:
- Players who have lost the session do not take rating points from each other. Only the winner takes rating points from other players. This means that victory is the only way to progress higher on the ranked ladder. If you didn’t win a session, you almost certainly will lose some rating points, even if you ended at the 2nd place.
- When you lose, you can mitigate the loss of rating points – but this doesn’t happen at the expense of other players. If you didn’t win, your final position doesn’t matter – instead, your final score matters. The closer you are to the leader, the fewer rating points you will lose (up to 0 if your score is equal and the victory was achieved by the second tie-break).
- The formula ignores the difference in rating that the players had prior to the session. It doesn’t matter if you are a strong player who has won against a weaker player, or vice versa. All that matters is how well each one of you performed in this particular session. This is because the matchmaking and the game itself can be quite random at times. We prefer not to have an extra punishment/reward for victories and losses that happened at the whim of luck. We want an environment where the only thing that matters is the player’s personal performance i.e., the score that was achieved during the session.
Rating loss by each loser = ( ːscoreː of the player - ːscoreː of the winner) / ːscoreː of the winner * 4 / number of players in the session. Rating gain by the winner = the sum of all rating losses.
[h3]What do we want to achieve with this change?[/h3]
- In sessions with 3 or more players, this eliminates any logical reason to play for any other place than the first one. If you cannot be the most prestigious gremlin in this session, you will lose prestige as the result. But even then, your best approach is to keep playing against the leader.
- This addresses the issues of 3-player sessions. When 2 out of 3 places are guaranteed to lose rating at the end of the session, the only viable strategy to avoid it for the 2nd and 3rd places is to play against the leader instead of playing against each other (which often happens under the current system).
- Sessions for rounds instead of sessions for points should become much more viable. The further the leader runs away with their high score, the bigger everyone’s loss of rating points gets. So it is in everyone’s interest to control the leader instead of hassling with each other for the 2nd or 3rd places.
- With these fixes, the available sets of rules for ranked sessions will increase without compromising with the quality of gameplay.
- Having a high rating would mean a lot! Since the only way to increase rating is to win sessions, players at the top of the ladder will prove that they are absolutely the best at winning, not just at seizing an opportunity of getting points at the 2nd place and minimizing their losses at the 3rd place. This will be true for all types of sessions, except for duels. Duels already are like that.
[h3]What are the possible issues with the new system?[/h3]
- If everyone but the winner loses rating points, ranked sessions might become somewhat discouraging. Success is hard to achieve and keep because anything but victory brings you down to a certain degree. Does this make the mode more competitive? Absolutely. More interesting? Mostly for those who are competitive, ready to play in such an environment, and most importantly are able to win often enough to move forward and raise their rating. But maybe this is exactly what the ranked mode should be: a competition among those who are interested in competing, and a competition for victory instead of a hassle for a higher place? We would love to put this theory in practice and see how this affects the game.
- Griefing might become a bigger issue because players who do not care about their rating will be able to cause more harm by attacking their personal target instead of the leader. On the other hand, revenge makes no sense in such an environment because pushing an opponent under yourself will not affect your own rating. And it is quite possible that griefers will fastly drop to the very bottom of the ladder simply because it will be hard to keep your rating high when you obviously aren't playing optimally.
- Griefing in duels. This issue is specific to duels. A player who is far ahead and has full control over the situation refuses to end the session and keeps playing for dozens of rounds just to boost their ego and/or humiliate the opponent? The new system might make such behavior more attractive, and technically it won't be against the rules. There are multiple ways to fix this, starting from the simplest: duels will be played for rounds instead of points, so each session would have an easily predictable end time. Other possibilities that are more sophisticated in design and implementation include the option for the players to surrender after a certain number of rounds has passed, or an option where sessions have two end conditions: the session ends if one of the participants reaches X points or after the fixed number of rounds.
As you can see, there are pros that can be very beneficial for the gameplay in ranked sessions, but there are also cons that have to be addressed in order to keep the game fair and interesting.
We definitely want to try the new formula for at least a few months, but we also are very interested to know your opinion about such a change: please share your comments here on Steam, or on the game’s Discord server, so that together we can create a better player experience.
Thanks for staying with us, and we look forward to the community discussion!
______
ːmaliceː Official Discord
ːmaliceː Twitter (game updates)
ːsummer_magicː Official Discord
ːsummer_magicː Twitter (game updates)